PCjr Memory Board (Racore / Radrive II, A2101 board)

[rvdbijl]

Hi Everyone!

I managed to snag a PCjr from a flea market earlier today and, a little to my surprise, it booted right up. Came with the IBM 4863 monitor and a spare keyboard as well.

Not having worked with or seen a PCjr up close ever (I have 2 5150's and a Compaq portable), I started the exploration to see what's in this box and whether or not everything works. This forum was great to get me started!

Based on what I've seen in this system, though, it looks like I have a Racore Radrive II on this machine. I started to test the components in this expansion unit, and the floppy drive works fine, and so does the real-time clock (after I replaced the battery). However, the memory is a different story. When booting up, the PCjr counts to 128K and then boots. There is, what looks like a memory card in the Radrive II expansion box, fully outfitted with RAM chips and labeled A2101-000. Doing some digging, this should mean I have 512Kb of memory on this thing..
When I look with memory dump tools, contents from 00000H to 1FFFFH are stable, but beyond 20000H it's a pile of junk that seems to have bytes flip constantly. So I'm guessing that something is wrong. I went so far as to pop the additional 256K of chips from their sockets, but the behavior remained the same.

Does anyone have any insights on what could be going on? I did a few simple tests, and it looks like the board is getting power, but I didn't go any further to look at the data lines, etc. I did remove the sidecar and refitted it, based on some posts from people saying that it could be a little iffy. No change there.. Pic of the board attached.

Any help is appreciated!

[Trixter]

First step is making sure the memory selector jumper didn't fall off :-) Documentation should be here: ftp://ftp.oldskool.org/pub/drivers/IBM/ ... re_ocr.pdf

Second step is probably to swap with known good RAM and see if that makes a difference. Couldn't hurt to reseat all the chips too, or maybe swap the banks they're in.

[Brutman]

Some of the Racore/Rapport expansion chassis have electrical problems - specifically if you over-tighten the sidecar screws it can cause breaks in the connectivity between the sidecar electronics and the RAM expansion board.

I would carefully inspect the system and look for things that are wiggling that should not be. Socketed chips should be firmly in their sockets. Connectors should be snug. Lightly tighten the sidecar screws .. finger-tight plus an eighth of a turn should be plenty for them. And of course all of this is static sensitive, so keep yourself properly grounded.

The edge card contacts can oxidize so just simply unseating and reseating them is often enough to make things work again. Use a pencil eraser and electrical contact cleaner for extra credit.

Generally these things are very reliable. A jumper falling out is unlikely.

[rvdbijl]

Haha! I think I tried all those things, and I did also play with the memory jumper on the board. It's there and it definitely did not fall out. Tried both positions, but no luck either way.

I'll pull apart the sidecar again later and take another peek to see if any components look loose. The RAM board has 1 bank soldered and another socketed. I removed the socketed chips and re-seated them, but no change in behavior. I did do plenty of wiggling with the sidecar and its connectors. From what I can tell, it all looks snug.

I also noticed the JrIDE thread.. That sounds like it would get me to 640K+ and add a little more functionality to this box..

Ultimately, it would be nice if I could get it connected to Ethernet as well... One step at a time...

[rvdbijl]

I spent a little more time with the PCjr and there is definitely something funky going on with the sidecar board.... If I leave it out of it's plastic enclosure and insert it directly to the side of the system, it's not recognized by its utilities and crashes. I suspect there are definitely a few traces on the board that didn't make it... But visually the board looks fine and nothing appears loose...
I think I'll aim for the JrIDE for a future upgrade instead of spending oodles of time trying to figure out which traces on the board have a fracture in them...

In the meantime the 2nd floppy and RTC do work.. As long as the board is in its plastic enclosure.

One more thing that I thought was curious. I was expecting the sidecar bus connector traces to be duplicated to the card edge connector, but going over some of the contacts with a multimeter, it's obvious that it's not. Or many traces are broken.

Pictures of the sidecar board attached.

[Brutman]

You've got a DMA capable unit there ... that was normally considered an upgrade as the PCjr did not come with DMA standard. But in practicality it is of limited use.

The edge connector only has to have enough pins to drive the memory card, and the memory card only needs a subset of the address lines present. You should not assume direct connections between the sidecar bus and that edge connector; it's not a bus extension.

[rvdbijl]

I realized it was a DMA unit when I saw the small ribbon cable in the manual. Still learning about the limitations of this PCjr system. Having always worked with 5150's and higher, I never got to play with one of these.
I do have to say the graphics and sound are miles beyond what CGA and PC speaker did on the normal PC/XT. Kings Quest I looks and sounds great on the PCjr. Amazing they could stuff all that in 128K.. Well.. I guess 96K since the video buffer must have been 32k in that mode.

Anyway -- Hoping Alan will start a new JrIDE production. When that's there, I'll pull out the RADrive and I'll probably sell it for parts to someone who wants to trace all the lines on the PCB.

[KenG]

I have the Quadram version of that with 384kb RAM, the clock and parallel port work. I replaced the 5.25 drive with a 720k 3.5. I don't use it anymore so I would be willing to sell for the right price.

[OldComputerGuy]

Why is there a dime in the battery holder? Maybe you're having an issue with the wrong type of battery.

[KenG]

I am fairly sure the battery only powers the clock and would not cause the issues he is having.